Training-free, Identity-preserving Image Editing for Fashion Pose Alignment and Normalization

Potito Aghilar, Vito Walter Anelli, Michelantonio Trizio, Eugenio Di Sciascio, Tommaso Di Noia

Abstract: Highlights•FashionRepose is a training-free pipeline designed for garment pose normalization.•The pipeline achieves identity-preserving edits, retaining logos and labels.•We propose a computer vision algorithm for identifying sleeve and torso regions.•The system employs a zero-shot methodology, without any finetuning or additional data.•FashionRepose offers a scalable and adaptable solution for the fashion industry.

External IDs:dblp:journals/eswa/AghilarATSN25